This invention relates in general to a computer management system and, more particularly, to automated management of an automatic call distributor center.
Current methods of monitoring a network, such as a call center with a plurality of switches, rely upon manual or human detection of problem conditions. Each call center is capable of simultaneously receiving and handling thousands of incoming calls into the call center. Each incoming call is routed by an automatic call distributor (ACD) to a call receiving agent. The call receiving agent can be an automated agent, such as an interactive voice response unit, or a person.
Call centers vary in size, from a few agents to several thousand agents. Typically, one or more supervisor work-stations are established to monitor the status of the agents and the calls received. Each supervisor work-station employs an individual (supervisor) to monitor a real-time agent status graphical display to visually watch for specific status patterns to detect failures of their respective agents. For example, a change in status of an agent could be represented by a change in color on the graphical display. Rapid detection of any failure relies on the supervisor's constant and astute attention. The supervisor must then determine and resolve the problem condition. However, human detection is ineffective and inefficient given that each supervisor may potentially monitor several call centers, with each call center having thousands of call receiving agents. Additionally, when the operator detects that the status of the call receiving agent has changed to a problem condition, the operator can not readily determine the cause of the problem.
Therefore, what is needed is a method and a system to automate monitoring of call centers for problem conditions that are detected based on defined parameters such that the cause of the problem condition is readily apparent from a triggered alarm condition.